VAN ALFEN, KOSUGE
Metabolism of the Fungicide 2,6-Dichloro-4-nitroaniline in Soil Neal K. Van Alfen*l and Tsune Kosuge
A 14C-labeledsample of 2,6-dichloro-4-nitroaniline (DCNA, Botran) was incubated in flooded soil that had been amended with D-glucose ( l % , w/w). After 3 days the amount of DCNA that could be recovered from the soil decreased to 7% and to 3% after 9 days. Evolution of radioactive carbon dioxide could not be detected even after the 9-day experimental period. Much of the radioactivity (ultimately 65%) could not be removed from the soil sample by extraction with an acetone-water solution. The major metabolite that was extractable from soil was identified as 4-amino-3,5-dichloroacetanilide (ADCAA). The expected metabolic precursor of ADCAA, 2,6-dichloro-p-phenylenediamine (DCPD), was also isolated from the extraction solution. A third compound isolated from soil appeared to be formed by the oxidative dimerization of DCPD. This same compound was also produced by the action of horseradish peroxidase on DCPD in vitro. Six fungi sensitive to DCNA were grown on potatwdextrose agar containing 10 pg/ml of DCNA, DCPD, or ADCAA. DCPD and ADCAA did not significantly affect the growth of any of the fungi tested as compared with controls whereas DCNA completely inhibited or reduced the growth rate in all cases.
Because 2,6-dichloro-4-nitroaniline (DCNA) is widely used for control of plant disease, knowledge of its fate in soil is important for its safe use as a fungicide. Experiments on its persistence in soil have revealed that breakdown, apparently as a result of microbial action, is more rapid in soil previously treated with DCNA (Groves and Chough, 1970). In the latter case, [14C]DCNAadded to soil was utilized rapidly and 25 to 50% of the radioactivity associated with the chemical was recovered in C02 evolved from the treated soil. When [14C]DCNA was added to soils not pretreated with DCNA, no 14C02 was evolved. However, it was not determined if DCNA had been altered without evolution of 14C02. Wang and Broadbent (1973) reported that only 2-4% of added DCNA was recovered after 10 days from glucose, alfalfa, and rice straw amended flooded soils as compared with 83.5% recovery from unamended flooded soils. From unamended upland soil, they reported that 99.5% of the DCNA was recovered unchanged after 4 weeks. In all the early studies, metabolic products of DCNA other than C02 were not identified and potential toxicity of the metabolites was not studied. Previously we reported that a number of bacteria and fungi in nutrient media have the capacity to convert DCNA to 2,6-dichloro-p-phenylenediamine (DCPD) and 4-amino-3,5-dichloroacetanilide (ADCAA) (Van Alfen and Kosuge, 1974). Since certain of the organisms that were most active in DCNA metabolism are common inhabitants of soil, we determined if similar DCNA metabolites were formed in soil. The purposes of this study were: (1)to identify the principal products of DCNA metabolism in soil and (2) to determine if any of the major metabolic products of DCNA were toxic to microorganisms. MATERIALS AND METHODS
Cultures. The cultures used in this study were obtained from the following sources in the Department of Plant Pathology, University of California, Davis: Teaching Collection, Alternaria alternata (Fr.) Keissler; Dr. E. E. Butler, Rhizopus oryzae Went et Prinsen-Geerligs (isolate number 1456),Rhizopus stolonifer (Fr.) Lind. (11361, and -
Department of Plant Pathology, University of California, Davis, California 95616. 'Present address: Department of Biology, litah State University, Logan, Utah 84322. 584
J. Agric. Food Chem., Vol 24, No. 3, 1976
Mucor hiemalis Wehmer (1237); Dr. R. K. Webster, Botrytis cinerea Fr. (417) and Mr. D. B. Marcum, Sclerotinia sclerotiorum (Libert) deBary. Chemicals. DCNA (2,6-dichloro-4-nitroaniline, Upjohn Co., Kalamazoo, Mich.) was recrystallized twice from benzene. Uniformly 14C-labeled DCNA, specific activity 0.98 pCi/pmol, was a gift of Dr. Joseph Ogawa, Department of Plant Pathology, University of California, Davis. Chromatography of the [ 14C]DCNAby silica gel thin-layer chromatography (TLC) in solvent a below showed it to be 99.7% DCNA. The remaining radioactivity (0.3%) did not move from the origin in the TLC system used. Synthesis of 2,6-dichloro-p-phenylenediamine (DCPD) and 4amino-3,5-dichloroacetanilide (ADCAA) was performed as previously reported (Van Alfen and Kosuge, 1974). Instrumentation. The required light absorption spectra were obtained on a Perkin-Elmer Model 337 and a Beckman ACTA I11 spectrophotometer, respectively. The mass spectra were obtained on an Associated Electronics Industries, Ltd. MS-7 and a Varian M-66 mass spectrometer. Radioactive counting was performed as previously reported (Van Alfen and Kosuge, 1974). Thin-Layer Chromatography. Thin-layer chromatography (TLC) was done on Eastman number 6061 silica gel plastic sheets or on Merck preparative silica gel plates in the following solvents: (a) benzene-diethyl ether (l:l, v/v); (b) benzene-petroleum ether (7:3, v/v); (c) chloroform-acetone (7:3, v/v); (d) ethyl acetate. Time-Course Study. Dry Yo10 loam soil (pH 8.6) untreated with pesticides for at least 3 years was obtained from a garden plot on the campus of the University of California, Davis. The soil was ground and mixed in a mortar and pestle. Portions (2.0 g) of the soil were placed in individual 10-ml beakers and flooded with 2.0 ml of heat sterilized 1%D-glucose solution containing 10 pg/ml of [ l4CC]DCNA.All operations were performed aseptically to avoid contamination by laboratory microorganisms. Each 10-ml beaker with soil was placed inside a 100-ml beaker which contained 15 ml of sterile 1 M KOH to trap C02. The 100-ml beaker was covered with filter paper and sealed with parafilm (American Can Co.). The efficiency of the system to trap COn was checked by release of 14C02 from an acidified solution of Na214C03 (100000 dpm, specific activity 5 pCi/pmol) and found to be essentially 100% effective. For controls, similar mixtures were prepared with soil that had been sterilized with propylene oxide. To
2,6-DICHLORO-4-NITROANILINE IN SOIL
sterilize the soil, 50 g was placed in a 1-pint jar humidified by a piece of moist filter paper, propylene oxide (2.0 ml) was added, and the jar was sealed for 6 days at 23 "C. The sterilized soil was amended with DCNA and glucose after the propylene oxide treatment. All of the samples were incubated a t 25 "C in the dark. Individual beakers of flooded soil were analyzed at regular intervals as follows: a sample of the KOH solution was examined in the liquid scintillation counter to determine if I4CO2 had been trapped. The soil was transferred to a test tube, 4.0 ml of acetone was added, and the mixture was agitated for 1 h on a New Brunswick twist action shaker. The mixture then was filtered through Whatman No. 42 filter paper and the soil on the filter paper was washed with an additional 4.0 ml of acetone. The filtrate and the acetone washings were combined and the extracted soil was air-dried. A weighed sample of the dry, extracted soil was finely ground and placed in radioactive counting solution with 0.4 g of Cabosil (Cabot Corp.) per vial to suspend the soil particles. Ten milliliters of a toluene-based scintillation counting fluid (Van Alfen and Kosuge, 1974) was added to each vial and the radioactivity in the samples then was determined in a liquid scintillation counter. The channels' ratio method was used to correct for quenching (Bush, 1963). In a control experiment, 88% of the DCNA mixed with soil was recovered by the extraction procedure. The acetone solution used to extract the soil sample was dried under a stream of N2 gas. The aqueous solution which remained was extracted with 5.0 ml of benzene. The benzene fraction was dried under a stream of N2 and the residue dissolved in 0.1 ml of acetone. The acetone solutions were chromatographed in solvent a or b. After development, the radioactive compounds on the chromatogram were located by autoradiography (Kodak No-Screen x-ray film). Portions on the chromatogram bearing radioactive compounds were cut out (plastic backed plates), individual pieces were placed in 5 ml of the toluene-based scintillation counting solution, and the samples were counted in the liquid scintillation counter. Large-Scale Isolation of Metabolites from Soil. Yo10 loam soil was amended with 100 pg of DCNA per g of soil and D-glucose (1%w/w) and then flooded with sterile water. After 5 days incubation in a 4-1. metal container, at 23 "C, the soil was extracted as described above, except that the mixture of soil-water with acetone was agitated on a rotary shaker, and the acetone-soil suspension and the benzene-aqueous solution ratios were, respectively, 1:l and 2:l (v/v). The benzene fraction was concentrated to 0.5 ml by evaporation, and chromatographed on Merck preparative silica gel TLC plates (2 mm thick), in solvent a. The desired metabolite was eluted from the silica gel with acetone and rechromatographed on Merck silica gel TLC plates (250 pm thick) in solvent c. The desired metabolites were eluted from the silica gel with acetone and the acetone solution evaporated to dryness. The residue was crystallized from acetone-water and the compounds analyzed by infrared spectrophotometry. Peroxidase Studies. DCPD was tested as a substrate for horseradish peroxidase (Sigma type 11),EC 1.11.1.7, by the procedure of Lieb and Still (1969). The reaction mixture was incubated for 10 min at 23 "C, and then extracted with 1vol of chloroform. The chloroform extract was chromatographed on Eastman silica gel TLC plates in solvent a. Toxicity of DCNA Metabolites to Certain Fungi. Difco potato-dextrose agar (PDA, 4% w/v) was amended
Table I. Distribution of Radioactivity into Various Fractions from [ 14C]DCNA Treated Soil
-_
Fractiona Days 0 1 1-Cc
2 3
3-c 4 9 9-C
Soil
Benzene
Waterb
Total
1300 11 300 1700 35000 50000 2400 54000 57 000 6700
120000 98000 102000 49000 39000 123000 39000 26000 107000
1400 3 500 1100 12700 15800 1600 14500 9200 1100
122700 112800 104800 96700 104800 127 000 107 500 92200 114800
a Values are given as total disintegrations per minute (dpm) in each fraction. Total dpm in the aqueous phase after benzene extraction. Distribution of the radioactivity of sterile controls after 1, 3, and 9 days is also indicated.
with 10 pg/ml of DCNA, DCPD, or ADCAA. The chemicals were first dissolved in 1.0 ml of acetone and then added to 500 ml of the medium after it had cooled but not yet solidified. An equivalent amount of acetone was added to unamended medium which served as the control. The media (30 ml) were then dispensed into petri plates (100-mm diameter) which had each been marked with four radii converging at the point where the inoculum was to be placed and inoculated with plugs (4 mm diameter) from the edge of actively growing fungal colonies on PDA. The petri plates were incubated at 25 "C in the dark and daily radial growth measurements were made along these radii. Each treatment was replicated three times. RESULTS
Time-Course Experiment. When the aqueous soil filtrates of the sterile controls and zero-time sample were extracted with benzene, 99% of the total extractable radioactivity as [14C]DCNAwas recovered in the benzene and 17'0 remained in the aqueous fraction. Distribution of the radioactivity in the other samples was a function of the length of the incubation period. Table I shows that the radioactivity extractable into benzene decreased rapidly with time of incubation of the soil. The 2-day benzene fraction contained less than one-half the amount of radioactivity that was present in the zero-time benzene fraction. The radioactivity bound to the soil increased rapidly and ultimately accounted for approximately 65% of the total detectable radioactivity. Throughout the 9-day experiment, no radioactivity was detected in the KOH trapping solutions, indicating that DCNA was not being metabolized to COz. As far as can be determined within the precision of the experiment, the data on total radioactivity recovered indicate that no major component is being lost from the extraction scheme as a function of time. Because these procedures would not differentiate between DCNA and its metabolites in the benzene fraction, the benzene extract was further purified by chromatography by the procedures described under Materials and Methods. In addition to DCNA, the chromatogram revealed four radioactive spots with R f values of 0.0, 0.25, 0.48, and 0.54. The spot with the Rfvalue of 0.54 had an intense pink fluorescence under uv light. The Rfvalues of two metabolites (0.25 and 0.46) were characteristic of ADCAA and DCPD, the previously reported microbial metabolites of DCNA. To quantitate the results, the amount of radioactivity associated with each spot then was determined. The results (Figure 1)show that as DCNA is metabolized, the radioactivity appears in several major fractions. As previously noted, the soil-bound counts increase most J. Agric. Food Chem., Vol. 24, No. 3, 1976
585
VAN ALFEN. KOSUGE
Table 11. Growth of Fungi in Media Amended with 2,6-Dichloro-4-nitroaniline or Its Metabolites Organism Controla DCNA~ DCPD
ADCAA
7.0 Botrytis cinerea 37.7 f 1.4 Rhizopus oryzae 35.4 f 1.4 R . stolonifer 14.8 i 0 . 6 Mucor hiemalis 17.9 2.6 Sclero tinia sclero tiorum 5.0 i 0.6 Alternaria alternata Measurements are given in millimeters of radial growth. Each figure given is the mean plus standard deviation from 12 The concentration of DCNA, DCPD, and radii. The measurements were taken after 24 h growth at 25 "C in the dark. ADCAA in the medium was 1 0 pg/ml. 5.9 f 0.5 37.4 i 0.9 40.1 i 0.5 14.0 5 0 . 3 18.5 i: 1.9 4.9 i 0.3
t
I 60 ' T .
-z- loor 3000
1 5 0 0 ~ ~ ' '10.00
, ,
,
400
6w
2025
I a
n
,
40
,
1
.
I
80 120 160 WAVELENGTH lMlCRONSI
60
200
I
240
Figure 2. Infrared spectrum of pink fluorescent compound isolated from soil. KBr pellet of sample was prepared for the spectrum.
20
2
4 6 8 T I M E (DAYS)
1
0
Figure 1. Distribution of radioactivity in various metabolites of [ 14C]DCNA from soil as a function of time. The radioactivity detected on the thin-layer plates was adjusted to correct for the efficiency of detection (approximately 45%): ( 0 ) DCNA; (0)soil-bound radioactivity; (m) ADCAA; (0)aqueous soil filtrate after benzene extraction; (A) pink fluorescent compound; ( A ) DCPD.
rapidly, followed by the rapid production of ADCAA appearing in the benzene fraction. The radioactivity that could not be extracted from the soil filtrate by benzene also increased significantly. At their maximum concentrations, the pink fluorescent compound and DCPD accounted for 5 and 3%, respectively, of the benzeneextractable radioactivity. Once the rate of metabolism of DCNA decreased, there was a slackening in the rate of increase of the metabolites. This was followed by a slight decrease in the amounts of all of the metabolites except the soil-bound one(s) which showed a slight net increase during this period. Identification of Metabolites i n the Benzene Fraction. To confirm that the spot with an R f value of 0.25 was [I4C]ADCAA, the silica bearing the labeled material was removed from the chromatogram, and extracted with acetone. Then a portion of the acetone sample was cochromatographed with synthetic ADCAA two dimensionally, first in solvent c (Rf0.37) and then in solvent d ( R f 0.39). The radioactive compound detected by autoradiography corresponded to the position of synthetic ADCAA on the chromatogram. A crystalline sample of the metabolite then was obtained from soil by large-scale isolation procedures as described under Materials and Methods. The infrared spectra of the soil metabolite and an authentic sample of ADCAA were identical. DCPD could not be isolated in amounts sufficient for ir spectrophotometry. However, the soil metabolite and synthetic DCPD chromatographed identically in solvents a and b. A pink fluorescent compound (Rf0.54) was also detected by the chromatographic procedures. Purification from soil was achieved by the procedure described under Materials and Methods with the following variation. After elution from a TLC preparative silica gel plate developed in 586
2opo , ,
:
E O L .
0
6.4 f 0.5 37.4 f 1.4 38.2 t 1.2 14.3 i 0 . 4 16.3 2.9 5.2 i 0.4
0 30.9 i 1.2 0 3.2 r 1.5 0 0
J. Agric. Food Chem., Vol. 24, No. 3, 1976
solvent a, the compound was adsorbed onto Bio-Rad alumina (neutral), activity grade I. The alumina sample was added to the top of a dry-packed column of the same type of alumina and the column washed with acetone. Fractions which contained the pink fluorescent metabolite were combined and then evaporated to dryness. The compound was crystallized from hexane as red needles. Since the compound decomposed above 300 "C, no melting point was determined. Mass spectrometry revealed a parent peak of mle 346 which had P + 2 and P + 4 peaks of 138 and 68% of the parent, indicating the presence of four chlorine atoms. The exact masses of the parent and P 2 peaks were 345.9348 and 347.9305 which correspond to a molecular formula of C12H~N4C14.The ir spectrum of the compound is shown in Figure 2. The uv-visible spectrum had a maximum at 290 nm and other peaks a t 380 and 510 nm. Peroxidase Experiments. When incubated at room temperature with horseradish peroxidase, the reaction mixture darkened rapidly and a dark-brown precipitate formed. After 10-min incubation the reaction mixture was extracted with chloroform which quantitatively removed the colored precipitate. Chromatography of the chloroform extract by TLC solvent a indicated that at least eight products were formed with the following R f values: 0.02, 0.06, 0.24, 0.45, 0.55 (pink fluorescence under uv light), 0.64, and 0.69 (one or more compounds remained at the origin). The compounds that were visually most abundant were the pink fluorescent compound ( R f0.55) and the R f 0.69 compound appearing dark red-brown in daylight. Varying the concentration of the substrate (DCPD) changed the relative abundance of these two compounds. The following reaction mixture gave the maximum production of the pink fluorescent compound: 0.7 ml of 0.2 M sodium acetate buffer (pH 5.1), 0.2 ml of 0.1% H202, 1.0 ml of enzyme preparation (10 pg/ml of Sigma type I1 horseradish peroxidase), and 1.0 mg of DCPD added to the reaction mixture in 0.2 ml of methanol. When DCNA replaced DCPD as the substrate, no reaction occurred. The pink fluorescent compound was extracted from the reaction mixture with 1 vol of benzene and the benzene extract was treated by the procedures described for the isolation of the pink fluorescing compound from soil. The
+
2,6-DICHLORO-4-NITROANILINE IN SOIL
ni 06 w 0 z
a
m
04
LL
w
m
a 02
0 200
300
400
500
WAVELENGTH Lnrn)
Figure 3. Comparison of uv-visible spectra of pink fluorescent compounds isolated from soil (A) and horseradish peroxidase reaction mixture (B). Methanol (S) was used as the solvent. The expanded spectra (A' and B') were run at a sensitivity of 0-0.1 as compared with 0-1.0 for the others.
absorption spectra of the pink fluorescent compound from the peroxidase reaction mixture and the DCNA metabolite from soil were identical (Figure 3) and both compounds chromatographed identically in solvents a and b. Toxicity of DCNA Metabolites to Certain Fungi. The relative toxicity of DCNA, DCPD, and ADCAA was tested against a number of fungi. The results (Table 11) show that although DCNA completely inhibited the growth of some of the fungi tested, and reduced the growth rate significantly in others, the two metabolites of DCNA did not significantly affect the growth of any of the fungi. In all cases, the radial growth of the fungi in the presence of DCPD and ADCAA after 24 h was essentially the same as the corresponding control. Daily measurements of radial growth for up to 4 days revealed no changes from the relative growth rates reported in Table 11. DISCUSSION
Experiments on the persistence of DCNA in soil have revealed that breakdown, apparently as a result of microbial action, is more rapid in soil previously treated with DCNA. Such results imply that pretreatment with DCNA causes the selection and buildup of microflora capable of metabolizing the chemical. In addition, certain soil amendments and low oxygen tension also promote utilization of DCNA by microflora. Thus, Wang and Broadbent (1973) reported that only 2-470 of added DCNA was recovered after 10 days from glucose, alfalfa, or rice-straw amended, flooded soil. Without amendment, 83.5% of the DCNA incorporated into the flooded soil was recovered. Such observations reveal that factors favoring activity of certain microflora largely determine the persistence of DCNA in soils. In our studies with glucose amended, flooded soil, DCNA is almost completely metabolized within 2 days to several compounds, two of which were not toxic to the microorganisms tested. The finding that DCPD is a metabolite of DCNA in soil is not unexpected. Our previous work (Van Alfen and Kosuge, 1974) has shown that a large number of fungi and bacteria in culture can rapidly reduce DCNA to DCPD, especially under anaerobic conditions. While the work of Briggs and Ogilvie (1971) would suggest that an acetanilide such as ADCAA should not accumulate in soil as we found, other studies using the herbicide metobromuron confirm that an acetanilide can accumulate as a metabolite of an aniline-based pesticide (Tweedy et al., 1970). Our results (Figure 1)indicate that ADCAA accumulates to a maximum at 4 days and declines in amount thereafter.
Since the pink fluorescent compound isolated from soil was the same as the one isolated from the DCPD-peroxidase reaction mixture, a similar oxidation of DCPD may account for its formation in soil. Wang and Broadbent (1973) also detected a pink fluorescent compound in flooded Columbia fine, sandy loam soil that had been amended with glucose and DCNA. The low amount of pink fluorescent compound relative to the amount of ADCAA in this experiment may perhaps be explained by the fact that oxidation would not be favored under the anaerobic conditions of flooded soil. Briggs and Ogilvie (1971) observed that soil slurries amended with 3chloro-4-methoxyaniline contained a pink compound which they isolated and identified as 3-chlorobenzoquinone4-(3'-chloro-4'-methoxy)anil.They also found that the latter compound was formed by a peroxidase-catalyzed oxidation of peroxidase 3-chloro-4-methoxyaniline. In addition, peroxidase catalyzes the formation of a tetrachloroazobenzene from 3,4-dichloroaniline, a soil metabolite of the herbicide Propanil (Bartha and Pramer, 1970). Although spectroscopic data are not sufficient to identify the structure of the pink fluorescent metabolite of DCNA, the accurate mass of the parent peak (345.9347) suggests a molecular formula of C12H6CldN4 (345.9348). Since it is known that this compound is formed by oxidation of DCPD, it is undoubtedly a dimer of DCPD. The lack of typical aromatic bands in the 2000-1660- or 900-700-cm-' regions together with the lack of the typical double absorption band of primary amines at 3500-3200 cm-' and C-N stretching of aromatic amines a t 1350-1250 cm-l support the evidence from the molecular formula that this compound is highly oxidized. The structure could be a double quinone diimine coupled by either an azine or mil. The intense band in the uv spectrum a t 290 nm and the compound's fluorescence would support such a structure. Although several other oxidative products of DCPD were formed in the peroxidase reaction mixture, none of these products was visually detected among the DCNA metabolites that were extracted from DCNA-amended soil. The composition of the microflora of the soil may be important in determining which metabolites of DCNA are formed and their relative abundance. The soil preparation procedure used in these experiments may also have affected the relative abundance of the metabolites since it has been observed that peroxidase activity is decreased by air drying of soil (Bartha, 1971). Experiments show that after prolonged incubation, a substantial portion of DCNA is converted to compounds that could not be extracted from the soil by our procedures. Although the compounds have not been identified, they may represent metabolites bound to soil particles (Hsu and Bartha, 1974). The radioactivity not extractable from soil was obviously not DCNA, since DCNA was readily extracted from all of the sterile controls as well as the zero time sample. ADCAA and DCPD appear to be detoxification produds of DCNA because they failed to inhibit growth of all the fungi and bacteria that were tested. In high nutrient soils that are subject to uncontrolled flooding, DCNA may be a poor choice for a fungicide since it is metabolized rapidly under those conditions. LITERATURE CITED Bartha, R., J. Agric. Food Chem. 19, 394-395 (1971). Bartha, R., Pramer, D., Adu. Appl. Microbiol. 13, 317-341 (1970). Briggs, G. G., Ogilvie, S. Y., Pestic. Sci. 2, 165-168 (1971). Bush, E. T., Anal. Chem. 35, 1024-1029 (1963). J. Agric. Food Chem., Vol. 24, No. 3, 1976
587
WIEDMANN, ECKE, STILL
Groves, K., Chough, K. S., J . Agric. Food Chem. 18,1127-1128 (1970). Hsu, T. S., Bartha, R., Soil Sci. 116, 444-452 (1974). Lieb, H. B., Still, C. C., Plant Physiol. 44, 1672-1673 (1969). Tweedy, B. G., Loeppky, C., Ross, J. A,, Science 168, 482-483 (1970). Van Alfen, N. K., Kosuge, T., J . Agric. Food Chem. 22,221-224 (1974).
Wang, C. H., Broadbent, F. E., J. Enuiron. Qual.2,511-514 (1973). Received for review September 15, 1975. Accepted January 22, 1976. This work comprises part of a thesis submitted to the University of California, Davis, California, by the senior author in partial fulfillment of the requirements for the degree of Doctor of Philosophy.
Synthesis and Isolation of 1-Hydroxy-2-propyl 3-Chlorocarbanilate from Soybean Plants Treated with Isopropyl 3-Chlorocarbanilate Jerome L. Wiedmann,* George G. Ecke, and Gerald G. Still The metabolism of isopropyl 3-chlorocarbanilate (CIPC) to form an alkyl hydroxylated metabolite in soybean shoots was indicated by alkaline hydrolysis-distillation of 3-chloroaniline from polar plant extracts. The presence of 1-hydroxy-2-propyl 3-chlorocarbanilate (111) in hydroponically grown plants was confirmed by autoradiograms of two-dimensional TLC plates, both of I11 and the acetate derivative of 111, t R of I11 on HPLC and t~ of the heptafluorobutyrate derivative of I11 on GLC. The presence of I11 in soil-grown plants was confirmed by the alkaline hydrolysis-distillation of 3-chloroaniline from polar extracts, t~ of the acetate of I11 on HPLC and GLC, and the mass spectral analysis of the acetate of 111. Soil grown soybean shoots produce a higher percentage of I11 when compared to the aryl hydroxylated metabolites which are the predominant metabolites found in hydroponically grown plants. Synthetic I11 (mp 88.5-91.0 "C, 68%) was prepared by HC1 cleavage of 1-triphenylmethoxy-2-propyl 3-chlorocarbanilate (V) (mp 138-139 "C, 82%) which was prepared from 1-triphenylmethoxy-2-propanol (IV) and 3-chlorophenyl isocyanate. Hodgson (1967) reported two water-soluble metabolites of isopropyl 3-chlorocarbanilate in barley, cucumber, and soybean. James and Prendeville (1969) reported the isolation of suspected @-glycosidicconjugates and postulated that the isopropyl moiety of CIPC was oxidized by smartweed, tomato, and parsnip plants. Extensive studies of hydroponically grown soybeans by Still and Mansager (1971-1973) showed the presence of the aryl hydroxylated metabolites, isopropyl 5-chloro-2-hydroxycarbanilate (I) and isopropyl 3-chloro-4-hydroxycarbanilate (11), but provided no indication of alkyl hydroxylation. Unpublished work at PPG Industries (1973) on soybean plants c o n f i e d the presence of I and I1 but also indicated the presence of a third metabolite. This paper reports the joint investigation resulting in the identification of this third metabolite, 1-hydroxy-2-propyl 3-chlorocarbanilate (111). EXPERIMENTAL SECTION
Synthesis of Metabolites. Compounds I and I1 were prepared by the method of Grunow et al. (1970). Preparation of the compounds by a different route has been described by Bobik et al. (1972). CH20H NHCOOC H (CH,),
I
VHCOOC H (C H ,)
I
I
-
NH COO; H
I11
OH
I1
Barberton Technical Center, Chemical Division, PPG Industries, Inc., Barberton, Ohio 44203 (J.L.W., G.G.E.) and Agricultural Research Service, US. Department of Agriculture, Metabolism and Radiation Research Laboratory, Fargo, North Dakota 58102 (G.G.S.). 588
J. Agric. Food Chem., Vol. 24, No. 3, 1976
In the following preparations, care was taken to use anhydrous solvents and reagents and to exclude atmospheric moisture from the reaction vessel. Compound IV (Figure 1) was prepared from chlorotriphenylmethane and 1,2-propanediol in the presence of pyridine by the method of Havbrandt et al. (1969); 86% yield of IV; mp 99-100 "C (lit. mp 96-97 "C). The use of purified chlorotriphenylmethane (Renfrow and Hauser (1943)) was found essential for obtaining high purity product. To a refluxing solution of 202.0 g (0.633 mol) of IV in 1000 ml of benzene was added 95.4 g (0.621 mol) of 3chlorophenyl isocyanate over a 25-min period. After refluxing for 14 h, unreacted isocyanate was destroyed by adding 50 ml of isopropyl alcohol and refluxing for an additional 15 min. The product, 1-triphenylmethoxy2-propyl 3-chlorocarbanilate, crystallized after the addition of 1500 ml of hexane. The product was removed by filtration, washed with 200 ml of hexane, and vacuum dried to yield 241.1 g (82%) of white crystals: mp 138-139 "C; ir (mull) 3310 and 1695 cm-l; NMR (CDCl3) 6 6.9-7.6 (m, 19, ArH), 6.72 (br s, 1, NH), 5.17 (m, 1, CH), 3.20 (d, J = 5 Hz, 2, CH2), 1.24 (d, J = 6.5 Hz, 3, CH3). Anal. Calcd for C29H26ClN03: C, 73.79; H, 5.55. Found: C, 73.52; H, 5.12. Hydrogen chloride was passed through a sintered glass bubbler into a solution of 240.4 g (0.510 mol) of V in 1200 ml of chloroform. The temperature rose from 24 to 28 "C over a 24-min period and then dropped to 24 "C during the subsequent 115 min of the addition period. About 20 g of hydrogen chloride was added during the first 50 min and only an additional 4 g during the remaining 89 min. After sparging the solution with nitrogen for 70 min to remove excess hydrogen chloride, the solvent was removed under vacuum using a 25-35 "C water bath. The 262 g of solid residue was dissolved in 450 ml of carbon tetrachloride and 93 g of crude 1-hydroxy-2-propyl 3-chlorocarbanilate was obtained upon cooling. The product was
